LUND UNIVERSITY
PO Box 117221 00 Lund+46 46-222 00 00
Fibroblasts as architects of cancer pathogenesis
Marsh, Timothy; Pietras, Kristian; McAllister, Sandra S.
Published in:Biochimica et Biophysica Acta - Molecular Basis of Disease
DOI:10.1016/j.bbadis2012.10.013
2013
Link to publication
Citation for published version (APA):Marsh, T., Pietras, K., & McAllister, S. S. (2013). Fibroblasts as architects of cancer pathogenesis. Biochimica etBiophysica Acta - Molecular Basis of Disease, 1832(7), 1070-1078. https://doi.org/10.1016/j.bbadis2012.10.013
Total number of authors:3
General rightsUnless other specific re-use rights are stated the following general rights apply:Copyright and moral rights for the publications made accessible in the public portal are retained by the authorsand/or other copyright owners and it is a condition of accessing publications that users recognise and abide by thelegal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private studyor research. • You may not further distribute the material or use it for any profit-making activity or commercial gain • You may freely distribute the URL identifying the publication in the public portal
Read more about Creative commons licenses: https://creativecommons.org/licenses/Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will removeaccess to the work immediately and investigate your claim.
1
Fibroblasts as Architects of Cancer Pathogenesis Timothy Marsh1, Kristian Pietras2, Sandra S. McAllister1,3
1 Hematology Division, Brigham & Women’s Hospital, Boston, MA 02115, USA 2 Department of Laboratory Medicine, Lund University Cancer Center, Lund, Sweden 3 Department of Medicine, Harvard Medical School; Harvard Stem Cell Institute; Broad Institute
of Harvard and MIT; Boston, MA 02115, USA Corresponding Author: Sandra S. McAllister Harvard Medical School Brigham & Women’s Hospital 1 Blackfan Circle, Karp 5-214 Boston, MA 02115 USA Email: [email protected] Phone: 617-355-9059 Fax: 617-355-9093
Abstract
Studies of epithelial cancers (i.e., carcinomas) traditionally focused on transformation of the epithelium (i.e., the cancer cells) and how aberrant signaling within the cancer cells modulates the surrounding tissue of origin. In more recent decades, the normal cells, blood vessels, molecules, and extracellular components that surround the tumor cells, collectively known as the “tumor microenvironment” or “stroma”, have received increasing attention and are now thought to be key regulators of tumor initiation and progression. Of particular relevance to the work reviewed herein are the fibroblasts, which make up the major cell type within the microenvironment of most carcinomas. Due to their inherent heterogeneity, plasticity, and function, it is perhaps not surprising that fibroblasts are ideal modulators of normal and cancerous epithelium; however, these aspects also present challenges if we are to interrupt their tumor-supportive functions. Here, we review the current body of knowledge and the many questions that still remain about the special entity known as the cancer-associated fibroblast.
2
1. Form and Function - Normal, Activated, and Cancer-Associated Fibroblasts
Fibroblasts are derived from the primitive mesenchyme, have an elongated, spindle-like
morphology, and are metabolically active (the suffix “blast” typically denotes an active
metabolism). As the most abundant cells of the connective tissue in animals, fibroblasts
both synthesize and degrade extracellular matrix (ECM) components by expressing
collagens, fibronectins, laminins, elastins, proteoglycans, integrins, matrix
metalloproteinases (MMPs), tissue inhibitors of metalloproteinases (TIMPs), and a host
of other ECM proteins that are expressed in a tissue-specific manner (reviewed in [1, 2]).
Consequently, fibroblasts are responsible for providing structural integrity to most
tissues. Fibroblasts also produce the tissue-specific basement membrane that provides a
protective barrier around the specialized epithelium, thereby contributing to specificity,
polarity, and functionality of the epithelium [2]. There is also evidence indicating that
fibroblasts communicate through the basement membrane to alter epithelial homeostasis,
proliferation, and differentiation [3].
Fibroblast activity is crucial during processes of wound healing and inflammation.
When tissue damage occurs, resident fibroblast populations proliferate and invade the
injured area in response to platelet clotting. Platelets adhere to exposed subendothelium
at sites of vessel injury and release their bioactive cargo (e.g., TGFβ1, PDGF, IL1-B,
MMPs, TIMPs), predominantly from α-granules that degrade the basement membrane,
induce cell proliferation and migration, and recruit inflammatory cells and fibroblasts
(reviewed in [4]). Under such conditions, fibroblasts are generally considered to be
“activated”. In particular, as the healing process progresses, fibroblasts turn on expression
3
of a filamentous actin, alpa-smooth muscle actin (αSMA), which enables them to exert
contractile forces to close the wound. Local tissue contractility is mediated by focal
adhesions between the activated fibroblasts – at this point called myofibroblasts – and the
ECM. Moreover, contractile fibroblasts are known to regulate interstitial fluid volume
and pressure via cytoskeletal infrastructure [5]. After wound closure, the balance of
MMPs and TIMPs secreted by fibroblasts is changed to favor ECM degradation (as
opposed to synthesis) which leads to massive apoptosis of the myofibroblast population.
Consequently, only quiescent, non-contractile fibroblasts are left at the resolved wound
site, and as such, myofibroblasts are only observed under pathological conditions.
The wound healing process seems to be co-opted by tumors; indeed, tumors have
been likened to “wounds that never heal” [6]. However, unlike wound resolution in
which fibroblasts “de-activate”, the myofibroblast population persists during fibrosis or
tumorigenesis for reasons that are not clear. It seems that normal fibroblasts have a
bimodal effect on cancerous cells in that early in tumorigenesis, fibroblasts work against
malignant progression, yet as the malignancy advances, fibroblasts are subverted to
promote tumor growth — these tumor-supportive fibroblasts are referred to as cancer-
associated fibroblasts (CAFs). In some cases, normal fibroblasts suppress malignant
conversion of immortalized prostate epithelium [7], whereas in the breast, normal
fibroblasts can induce the transition of already transformed ductal carcinoma in situ to
invasive carcinoma [8]. The oncogenic transformation of the epithelium may subvert
normal fibroblasts and potentiate their ability to promote tumor growth. Concordantly,
one study has shown that suppression of the retinoblastoma (Rb) protein in pancreatic
epithelium induces a selection pressure for fibroblasts that lack p53 and subsequently
4
results in p53-inactivated epithelium [9]. Although the reasons why CAFs remain
perpetually activated remain to be elucidated, it is very clear that fibroblasts participate in
an elaborate, reciprocal cross-talk with the cancerous epithelium. (this was redundant
with above)
2. Cancer-Associated Fibroblasts - Heterogeneity or a Spectrum of Phenotypes?
It is widely accepted that CAFs are a heterogeneous cell type and that this diversity may
arise from their cell(s) of origin, the tissue in which they develop, or their activation state
at any given time. This heterogeneity has presented challenges to precisely and
exclusively identifying CAFs and to distinguishing them from other cell types that
express similar markers upon histopathological analysis of tumors and tissues (Figure 1).
Instead, CAFs are more readily distinguished from their normal counterparts by their
phenotype, proliferation rate, and differential expression of ECM constituents [10].
CAFs are most often denoted by expression of αSMA. Several additional markers
are used to identify CAFs, including: vimentin, platelet-derived growth factor receptor
alpha (PDGFR-α), platelet-derived growth factor receptor beta (PDGFR-β), fibroblast
specific protein (FSP-1), and fibroblast activation protein (FAP) [11-14]. Nevertheless,
no one marker specifically labels all CAFs or clearly distinguishes CAFs from normal
fibroblasts or other closely-related cell types. These other cell types include pericytes
(cells that line blood vessels, also known as mural cells), smooth muscle cells, epithelial
cells that have undergone an epithelial-to-mesenchymal transition (EMT), myoepithelial
cells (specifically in the breast), and some adipocytes (Figure 1). Most often, in order to
5
generally classify these various cell types, a combination of markers must be used. For
example, αSMA-positive CAFs can be distinguished from pericytes, which stain
positively for neuron glial antigen 2 (NG2) and regulator of G-protein signaling 5
(RGS5). RGS5 has been shown to be overexpressed in abnormal tumor vasculature and
colocalizes predominantly with PECAM-1/CD31 and less so with PDGFR-α and αSMA
[15]. Although some carcinoma cells express FSP-1, FSP-1-positive fibroblast sub-
populations present in the tumor microenvironment have been shown to facilitate
malignant progression. For example, in a syngeneic mouse model of melanoma, PDGF-
CC signaling recruited fibroblasts with differential expression of FSP-1, PDGFR-α and
αSMA [11]. Additionally, vimentin is expressed in most mesenchymal cell types as well
as epithelial cells that have undergone an epithelial-to-mesenchymal transition (EMT).
Due to the apparent heterogeneity of fibroblasts and their diverse origins, it has therefore
been difficult to distinguish true fibroblasts from fibroblast-like cells. Moreover,
identifying markers to label fibroblast sub-populations that exclusively contribute to
cancer progression in various organs has presented challenges (Figure 1).
Molecular profiling studies have also revealed the heterogeneity of fibroblast and
CAF populations, yet have also suggested that core signatures, at least among sup-
populations of fibroblasts, might predict tumor-supportive function. For example, gene
expression analysis of fibroblasts isolated from breast cancer patient tumors yielded
subtype-specific molecular signatures, especially with respect to expression of genes
encoding cytoskeletal and integrin signaling proteins [16]. On the other hand, a study in
which fibroblasts were isolated from ten different anatomical regions and exposed to
serum (mimicking a wound response), revealed a common transcriptional signature,
6
termed the fibroblast core serum response (CSR), that was also identified in CAFs
isolated from various carcinomas and predicted metastatic progression in patients with
breast, lung, and gastric cancers [16]. Similarly, differences in tumor-promoting ability
were found between normal tissue fibroblasts and CAFs when examined for their
prostaglandin (PGE2) secretory phenotype, which is elevated in tumors [17]. Two recent
studies defined very similar CAF expression profiles that represented pro-inflammatory
signatures also found in CAFs derived from cancer patients. In one study using a K14-
HPV16 mouse model of multistep squamous skin carcinogenesis, this signature included:
Cox2, IL-1β, OPN, IL-6, CXCL1/2 [18]. In the other study using a xenograft model of
breast cancer progression, enhanced expression of many of these same proteins were
found in CAFs relative to normal mammary fibroblasts [19]. Importantly, this second
study also identified the molecular modulator that caused fibroblasts to adopt this pro-
tumorigenic CAF signature – the secreted growth factor, granulin (GRN) [19]. Hence,
common biological responses of fibroblasts to their microenvironmental cues (e.g., serum
exposure) might reveal how fibroblasts acquire their CAF phenotypes. However, these
responses seem restricted to different subpopulations of fibroblasts. Given this diversity
of biological functions and their obvious heterogeneity, markers and methods to identify
different CAF populations for therapeutic purposes, while challenging, would seem of
utmost importance.
3. Fibroblasts in Cancer Pathophysiology
7
It has long been thought that fibroblast behavior is dictated by the epithelium, but
recently more attention has been paid to the possibility that fibroblasts actively drive
tumorigenesis and cancer progression [8-11, 20, 21]. There is now evidence to suggest
that fibroblasts play important roles during the entire course of tumor development, from
the pre-neoplastic state until the terminal stage of cancer progression - metastasis.
3.1 Cancer Initiation – Do Fibroblasts Direct Tumorigenesis?
Tumor initiation is typically conceptualized as the accumulation of genetic and epigenetic
mutations in the epithelium that results in recruitment of a reactive stroma. While the role
of fibroblasts in de novo transformation or induction of carcinoma from epitlehlum
lacking oncogenic mutation is currently debated, some studies have shown that
fibroblasts facilitate carcinoma formation from epithelium that is cancer-prone.
Studies of prostate cancer have demonstrated that isolated CAFs, but not normal
fibroblasts, can induce the transformation of immortalized epithelial cells [20, 22].
Transgenic mouse models have provided some insights into CAF-derived factors that are
responsible for tumor initiation. For example, Wnt1 overexpression in fibroblasts
transforms mammary epithelial cells from C57BL/6 mice [23]. Additionally,
overexpression of HGF and/or TGFβ in fibroblasts was demonstrated to be sufficient for
inducing ductal carcinoma in situ (DCIS), adenocarcinoma, and poorly differentiated
tumors in the breast [24]. Knockout models and depletion experiments have also
demonstrated the importance of fibroblast activation in tumorigenesis. One study using
FSP-1-deficient mice showed reduced tumor growth and attenuated metastatic potential
of an otherwise highly metastatic murine mammary carcinoma cell-line, whereas
8
injection of wild type fibroblasts partially rescued this effect [25]. Furthermore,
knockout of TGFβRII in FSP-1-positive cells promoted prostate neoplasia and
forestomach squamous cell carcinoma [10].
A recent study using mice containing conditional alleles of Pten and an Fsp-cre
transgene, showed that inactivation of PTEN specifically in mammary fibroblasts
significantly increased the incidence and rate of progression to adenocarcinoma of
MMTV-ErbB2/Neu-driven tumors [21]. Upon examination of the pre-neoplastic
mammary glands of the mice in this study, significant increases in ECM remodeling and
immune cell infiltration were observed. Furthermore, the PTEN-specific gene signature
of the pre-neoplastic fibroblasts from these mice was remarkably similar to stromal
signatures found in patients with breast cancer. However, as mentioned earlier, FSP-1
marks both normal and activated fibroblasts, as well as some epithelial cells, in the
mammary gland [26], so it is not clear that the accelerated growth was due exclusively to
the PTEN-null fibroblasts.
Finally, a number of studies have implicated reactive stroma, including activated
fibroblasts, in accelerating the appearance of carcinoma (reviewed in [27]). Many of
these studies incorporate the use of irradiated fibroblasts with pre-transformed epithelium
that bear mutations in tumor suppressor genes. The most recent example comes from a
study in which Trp53-deficient mammary epithelilum was transplanted into irradiated
hosts (whole body irradiation) that had previously undergone mammary fat pad clearance
[28]. In this study, estrogen receptor-negative tumors developed with increased
incidence and at an accelerated pace above controls. Another type of reactive fibroblast,
the senescent fibroblast, has also been implicated in driving cancer progression from pre-
9
malignant epithelium. A recent study using a model of skin carcinogenesis elegantly
showed that senescent fibroblasts drove progression of pre-neoplastic epithelium in an
osteopontin-dependent manner [29].
3.2 Early Stage Cancer – Fibroblasts Pave the Way for Cancer
There is clear evidence that CAFs play an important role during the early stages of
tumorigenesis. As tumors progress, the architecture of the host tissue becomes highly
distorted with aberrant accumulation of ECM components. One hallmark of this process
within the tumor microenvironment is collagen cross-linking. Hints about the molecular
basis of this come from experimental models showing that CAFs are a major source of
lysyl-oxidase (LOX), which cross-links collagen, and that inhibiting LOX delays the
onset of breast cancer [30]. In fact, LOX activity has been observed in many types of
aggressively growing cancers [30]. Likewise, in breast and colon cancer models, CAFs
express uPA and uPAR, which modulate ECM degradation, which, in turn, increases the
bioavailability of growth factors that are typically sequestered by the ECM [31].
In addition to modulating ECM components, CAFs are known to regulate
proliferation of the epithelium. In some cases, enhanced proliferation is the result of
direct paracrine interactions between tumor cells and CAFs, and in others, it appears to be
an indirect effect of inflammatory processes in which CAFs serve as mediators (discussed
in more detail later). Under various contexts, CAFs produce a repertoire of growth factors
and cytokines that influence the behavior of the epithelium, including HGF, EGF, IGFs,
IGFBPs, b-FGF, TGFβ, to name a few (reviewed in [20]). In other contexts,
inflammatory processes that impinge upon fibroblasts endow them with the ability to
10
directly affect tumor cell behavior. For example, generation of reactive oxygen species
(ROS) in the inflammatory microenvironment has been shown to modulate CAF
transdifferentiation, thereby enhancing their tumor-promoting functions [32, 33].
Interestingly, it has also been shown that induction of autophagy in the CAF
compartment helps to promote tumor cell survival via processes involving
downregulation of caveolin-1 and the subsequent stabilization of hypoxia-inducible
factor-1 alpha (HIF1-α) [34].
3.3 CAFs in Cancer Progression and Metastasis
Metastasis, which is the leading cause of death in patients with solid tumors, is a highly
inefficient process, and it is estimated that less than 1% of disseminated tumor cells are
able to form viable metastases. Not surprisingly, successful metastatic progression also
involves fibroblasts. Within the primary tumor site, it has been shown that factors
secreted by CAFs, including TGFβ-1, induce tumor cells to undergo EMT, thereby
promoting tumor cell motility, invasion, and metastasis [35, 36]. Likewise, knockout of
one allele of TGFβRII in FSP-1-positive cells yielded more metastases in the MMTV-
PyMT mammary tumor model [37]. Underscoring these results, transgenic mice deficient
in FSP-1 cells did not incur metastases when engrafted with a highly metastatic murine
mammary carcinoma cells [25]. As primary modulators of the ECM, as mentioned
earlier, CAFs certainly help tumor cells to invade the surrounding tissue by forming the
invasive front [38]. Indeed, it was demonstrated that CAF expression of caveolin-1 leads
to matrix remodeling, invasiveness, and increased metastases in mice injected with breast
adenocarcinoma cells [39].
11
Fibroblasts likely play a role at metastatic sites as well, as activated fibroblasts
have been observed at metastatic sites where they promote tumor cell proliferation [40] .
Fibroblast production of periostin at the site of a micrometastasis was shown to be crucial
for metastatic colonization and maintenance of cancer stem cell-like properties, which are
thought to be necessary for initiation of a new tumor [41]. Other studies have shown that
systemic signaling cascades that operate as a consequence to the body’s response to
cancer, aid construction of the tumor-supportive microenvironment, including CAFs, at
metastatic sites [19, 42]. Such studies have raised important questions that have yet to be
fully elucidated, such as: what is the source of CAFs found in metastatic sites? Do
disseminated tumor cells have to “start all over again” to build their specialized CAF
compartment? One early study suggested the interesting possibility that tumor cells take
their stroma, including CAFs, with them [43]. The stromal requirements of
micrometastases is a rate limiting step in the colonization of secondary organs and fully
elucidating the mechanisms by which normal stromal compartments of the secondary
organ prevent or promote metastatic growth is crucial to developing therapies against this
life threatening condition.
4. Tracing the Origins of CAFs
Perhaps one of the most asked questions in the study of tumor microenvironment is:
what is the source of cancer-associated fibroblasts? It is likely that CAFs are derived
from a variety of sources. The origin of myofibroblasts in fibrosis has been extensively
studied and has provided insight, both technically and conceptually, into the source of
12
CAFs in cancer. Currently, there are three prevailing hypotheses regarding the source of
activated fibroblasts in tumors. The first, and most commonly studied, is that CAFs
derive from resident cells that are recruited from local sources into the tumor
microenvironment by the malignant epithelium. These resident cells might include
normal tissue fibroblasts (as discussed in detail above), and tissue mesenchymal stem
cells (MSCs). Second, circulating fibrocyte progenitors and bone marrow-derived cells
have been demonstrated to extravasate into the tissue where the tumor resides and
differentiate into cells with fibroblast-like phenotypes. A third and final scenario is that
other cell types, such as pericytes, myoepithelial cells, and endothelial cells
transdifferentiate to give rise to the tumor fibroblast-like population.
4.1 Calling upon Resident Fibroblasts
As discussed at length earlier, several lines of evidence suggest that activation of resident
fibroblast populations give rise to CAFs. Studies of the origin of fibroblasts during
wound healing and organ fibrosis have helped shed light on how this process occurs in
cancer. As stated earlier, the tumor and fibrotic microenvironment is a key regulator of
pathological progression. Resident fibroblasts surrounding the malignancy are commonly
thought to be the first responders to the site of insult that the tumor provides.
Although not well understood, resident mesenchymal stem/progenitor cells
(MSCs) have been shown to contribute to tumor growth in various model systems [44],
suggesting that these cells, which share a lineage relationship with fibroblasts, might
provide a source of CAFs. MSCs have been shown to differentiate into myofibroblast-
like cells (marked by αSMA and collagen 1) [45] and to secrete tumor-promoting factors,
13
including VEGF, IL-8, HGF and IGF-1 [46]. It is not clear whether resident MSCs
differentiate into CAFs for several reasons. First, techniques to isolate MSCs that can
differentiate into the various lineages from resident tissue is challenging and it is not clear
which cell-surface markers would clearly and exclusively distinguish MSCs, if possible.
Second, once MSCs are brought into cell culture, they are highly susceptible to
differentiation and selection pressures.
4.2 Bone Marrow and Circulating Cells as CAF Reservoirs
As vasculature in tumors is highly permeable, the influx of circulating cells increases and
the relative contribution of bone marrow-derived cells to the tumor stroma also increases.
The contribution of bone marrow-derived cells to the CAF population is debated and
appears to be context-dependent. In a study of inflammation-induced gastric cancer, up
to 20% of the αSMA+ in the tumor stroma were shown to be derived from bone marrow
mesenchymal progenitor cells [47]. Yet studies of pancreatic and other cancers
demonstrated that bone marrow-derived cells contribute a small percentage of the CAF
population [48, 49]. A study of breast cancer, in which tumors formed with a highly
reactive, myofibroblast-rich stroma, demonstrated that none of the αSMA+ population
was of bone marrow origin [19]. However, interesting studies in patients who had
previously received a bone marrow transplant from a donor of the opposite sex indicated
that some tissue fibroblasts were of donor origin [50, 51].
Bone marrow-derived MSCs have been of interest due to their ability to home to
sites of inflammation, tissue repair, and neoplasia in experimental models [52-54]. Other
studies have identified some of the growth factors and cytokines that attract MSCs to
14
tumor sites (e.g., VEGF, EGF, HGF, b-FGF, PDGF, CCL2) [55-57]. In studies using an
ovarian cancer cell-line, Skov3, admixed with human MSCs, the resulting tumor
microenvironment harbored abundant cells with both CAF and pericyte-like phenotypes
[57]; however, this study did not determine whether the CAFs arose from the implanted
MSCs or from recruitment of other host cells that gave rise to the CAF population. Bone
marrow-derived MSCs have also been shown to play an important role in tumor
progression [58]. Nevertheless, whether or not MSCs should be equated with CAFs is
not yet clear.
Circulating fibrocytes have also been considered as a source of CAFs, as they
have been shown to be recruited to injured tissues [59, 60] and to contribute to the αSMA
myofibroblasts observed in a mouse model of allergic asthma [61]. Furthermore, in a
model of bleomycin-induced lung fibrosis, GFP bone marrow chimera mice have
elevated levels of GFP+, collagen 1 producing fibroblasts in the lungs [62]. Fibrocytes
are thought to be of bone marrow origin due to their expression of leukocyte marker
CD45, bone marrow stem cell antigen CD34, CD11b, and fibroblastic markers such as
vimentin, collagen I/III and fibronection (reviewed in [63]). As fibrocytes can loose
CD45 and CD34 markers when in circulation it is likely that the cummulative effects of
fibrocytes in tissue fibrosis and cancer is unrecognized.
4.3 Transdifferentiation of Resident Tissue Cells
The plasticity of various epithelial and mesenchymal cell types during pathological
processes has become a topic of intensive investigation and discussion. A number of cell
types found in the tumor microenvironment or in proximal tissue have been proposed as
15
candidates for the CAF compartment. These cells include: pericytes, adipocytes,
endothelial cells, and even the epithelial cancer cells.
Pericytes, which inherently express αSMA, have been of particular interest as a
speculated source of fibroblasts in recent studies, mainly due to their reported functions
in other pathological conditions. One study utilized lineage tracing of
pericytes/perivascular cells to demonstrate their transdifferentiaton into interstitial,
proliferative myofibroblast-like cells in kidney fibrosis [64]. Additionally, ADAM12+
perivascular cells in fibrotic regions have been shown to undergo a progressive
differentiation program into myofibroblasts [65]. Furthermore, pericytes have been
shown to contribute to scar formation in spinal cord injury [66]. Pericytes are known to
detach from tumor vasculature, thereby contributing to the inherent “leakiness” of tumor
vasculature [67] [68], leading to the idea that these pericytes, should they remain viable
in the tumor microenvironment, might give rise to cells with a CAF-like phenotype.
Nevertheless, whether or not pericytes contribute in a major way to the CAF popualation
remains to be determined conslusively.
Given their mesenchymal lineage relationship with fibroblasts, tissue adipocytes
have been proposed as a likely source of CAFs. For example, in a breast cancer model,
local adipose tissue contributed ~29% of the αSMA-positive population and comprised
~27% of the NG2-positive cells [69]. Moreover, adipose precursor cells have been shown
to induce expression of fibronectin, αSMA, and vimentin in 4T1 murine mammary
carcinoma cells, consistent with tumor-associated fibroblastic cells [70]. Finally, one
study demonstrated that breast tumor-derived TNF-α and IL-11 prevented differentiation
of adipocyte precursors, causing them to expand as a fibroblastic population to contribute
16
to the desmoplastic stroma [71]. Nevertheless, direct proof that mature adipocytes de-
differentiate within the tumor microenvironment (e.g., in the mammary gland) to give
rise to the CAF population has not been provided as of yet.
One lineage tracing experiment in a model of kidney fibrosis demonstrated that
endothelial cells acquire fibroblast markers during a processes coined “EndMT” –
endothelial to mesenchymal transition [72]. In this context, endothelial cells underwent a
partial transdifferentiation to acquire fibroblastic markers αSMA, FSP-1, vimentin, and
N-cadherin, while retaining endothelial markers VE-Cadherin, CD31, Tie1, and Tie2
(reviewed in [73]). A related study showed that up to 40% of CAFs arise as a
consequence of EndMT in two different murine cancer models [74].
Perhaps the most accepted transdifferentiation program occurs during embryonic
development when epithelial cells transition into mesenchymal cells, in a process termed
the epithelial-to-mesenchymal transition (EMT). The existence of a permanent and
irreversible EMT in adult tissue is often debated. Recently, the EMT has been classified
into three functionally different processes that occur in distinct biological settings [35].
Type 1 which occurs during implantation, embryogenesis, and organ development is well
studied and will not be discussed here (reviewed in [35, 75]). Type 2 occurs during tissue
regeneration and fibrosis (discussed by other authors in this special issue). Type 3,
discussed below, is associated with tumor progression and metastasis. As discussed
earlier, in general, the cancer research community views EMT as a process by which
cancerous epithelial cells undergo a partial, and possibly reversible, transition to a
mesenchymal-like state for the puposes of invasion and metastasis. Epithelial cells that
undergo EMT in vitro are marked by acquisition of mesenchymal markers (e.g., αSMA,
17
FSP-1, vimentin, desmin, and N-cadherin) and loss of epithelial markers (e.g., E-
cadherin), while for the most part, retaining expression of epithelial-specific cytokeratins
(reviewed in [75]). It has been suggested that epithelial cells undergo an EMT after
exposure to oxidative stress induced my MMPs to become myofibroblasts [76]. The
EMT phenotype therefore supports functions that are not normal to terminally
differentiated epithelial cells, such as anchorage-independent survival, loss of homotypic
cell-cell contact, cell motility, invasion, and ability to breach the basement membrane
(reviewed in [35, 77, 78]), several properties that are characteristics of fibroblast-like
cells. Indeed, an epithelial cell-line derived from a breast cancer biopsy displayed
fibroblast characteristics and had the ability to differentiate into myofibroblasts [79].
Cancer cells that have undergone EMT are typically found at the invasive front of tumors
in various model systems, further supporting their role in the metastatic process. At this
point, while it is generally thought that the EMT is important for cancer progression, it is
still debated whether cells that have undergone an EMT fill the role of CAFs.
The underlying heterogeneity of CAFs and the promiscuity of markers that are
used to identify them (Figure 1) also present challenges to identifying their origins.
Additional studies using transgenic models and in vivo lineage-tracing techniques that
incorporate precise, well-accepted, cell-specific promoters might help us find more
answers to the questions about the origins of CAFs.
5. Interactions in the Tumor Microenvironment
18
Crosstalk within the tumor microenvironment is not limited to paracrine signaling
between CAFs and malignant cells, but also occurs between different resident and distant
stromal cell types. CAFs actively engage these multiple other stromal cell types,
including endothelial cells, BDMCs and inflammatory cells, in order to stimulate cancer
growth, angiogenesis and metastatic spread.
5.1 CAFs and Endothelial Cells
Cancer-associated fibroblasts participate in processes supporting tumor growth,
angiogenesis and progression through multiple and principally different mechanisms
involving crosstalk with other cellular compartments within the tumor microenvironment.
In particular, CAFs have been highlighted as providers of a multitude of pro-angiogenic
cues (growth factors, proteases, extra-cellular matrix constituents, cellular recruitment) in
various tumor types. The stimulation of angiogenesis afforded by stromal fibroblasts is in
line with their primary localization at the leading edge of tumors, where there is a
manifested demand for an expanded vascular supply [38, 80]. CAFs secrete and deposit
numerous growth factors in the tumor microenvironment that stimulate endothelial cell
growth and angiogenesis. Studies using mice genetically engineered with a reporter for
vascular endothelial growth factor (VEGF)-A have demonstrated dramatic induction of
VEGF-A transcription in the stroma of both spontaneously arising, as well as implanted
tumors [81]. Indeed, stromal cells in ovarian carcinomas provide most angiogenic growth
factors in higher quantities than do the overt malignant cells [82]. Provision of
angiopoietin-1 and -2 by CAFs acts to stabilize the neo-vasculature of ovarian
carcinomas [83], indicating that fibroblasts in this context also stimulate vessel patency.
19
While neo-angiogenesis is typically initiated by the hypoxic tumor
microenvironment, CAFs are also in many cases induced to secrete pro-angiogenic
factors in response to paracrine signaling events. Production of various isoforms of the
platelet-derived growth factor (PDGF) family by malignant cells serves to recruit and
activate pro-angiogenic CAFs in various cancers [84]. Hence, paracrine PDGF
stimulation of CAFs induces production of the prototypical pro-angiogenic inducer
fibroblast growth factor (FGF)-2 in both cervical carcinomas and melanoma [11, 85].
Interestingly, CAFs activated by PDGF-CC in melanomas also secrete the
extracellular matrix protein osteopontin, the action of which is known to synergistically
stimulate angiogenesis together with FGF-2 and promote autocrine VEGF-A signaling in
endothelial cells [11, 86-88]. Consistent with the role of PDGFs as essential upstream
mediators of angiogenic cues provided by CAFs, malignant cells genetically deficient for
VEGF-A make up for this lack by paracrine activation of PDGFR-α in CAFs, thus
stimulating production of stromal VEGF-A [89, 90]. Moreover, autocrine stimulation of
CAFs by the cytokine CXCL14 in prostate cancers also results in induction of FGF-2 and
subsequent neo-angiogenesis [91].
Finally, CAFs indirectly regulate the process of angiogenesis through abundant
provision of extracellular matrix (ECM) products, such as collagen(s), osteopontin and
tenascin-C, as well as by secreting matrix-remodeling proteases, including members of
the matrix metallo-proteinase (MMP) and the a disintegrin and metallo-proteinase
(ADAM) families [92]. Paradoxically, excessive production of ECM and a prolonged
fibrotic reaction is detrimental to the angiogenic process, most likely by physically
restricting endothelial sprout formation and tip cell migration, and attenuation of the
20
desmoplastic response in pancreatic adenocarincomas through inhibition of hedgehog
signaling alleviates the angiogenic blockade and improves tumor perfusion [93, 94].
Taken together, our current knowledge points towards a pivotal role for stromal
fibroblasts in releasing tumors from dormancy through activation of the angiogenic
switch, and identifies CAF-endothelial cell crosstalk as a hitherto unexploited target for
cancer therapy.
5.2 CAFs and Bone Marrow-Derived Cells
In addition to supporting tumor growth through orchestration of paracrine signaling
networks within the tumor parenchyma, CAFs mediate long distance effects through
mobilization of various bone marrow-derived cells (BMDC). Strikingly, CAFs, but not
normal fibroblasts, secrete stromal cell-derived factor (SDF)-1 in order to recruit
Sca1+CD31+ BMDCs that act as endothelial progenitors and incorporate into the neo-
vasculature [95]. In this study, SDF-1 was specifically expressed by alpha-smooth muscle
actin (αSMA)-positive myofibroblasts.
In addition to CAFs affecting BMDCs, BMDCs can affect the CAF phenotype.
Interestingly, recruitment of a distinct set of Sca1+c-Kit- hematopoietic BMDCs induces
myofibroblast differentiation and expression of αSMA through secretion of granulin
(GRN) [19]. The mature myofibroblast component subsequently promotes multiple
aspects of tumor growth, angiogenesis and progression systemically through production
of an array of cytokines and growth factors, including CXCL1, CXCL2, IL1A and B, IL6
and IL8 [19]. Thus, reciprocal signaling between CAFs and BDMCs represents a
21
conceptually novel and viable cancer drug target, e.g., through targeting of SDF-1 or
GRN.
5.3 CAFs and Immune Modulation
It is increasingly appreciated that phenotypic modulation of the immune response is a
feature of most, if not all, cancers [96]. CAFs manipulate the inflammatory
microenvironment through two principally distinct mechanisms. Firstly, CAFs harbor a
pro-inflammatory expression profile that serves to recruit macrophages, neutrophils and
other stimulatory immune cells that promote various aspects of tumor progression [19,
97-99]. Notably, CAF-derived S100A4 mediates tumor infiltration of T-cells, the action
of which promotes lung metastatic formation in experimental breast cancers [100].
Secondly, immune editing by CAF-derived factors act to suppress tumor detection
and rejection by the host immune system [101]. In this context, stromal fibroblasts from
melanoma, but not from normal skin, impede NK cell cytotoxicity both through cell-to-
cell contacts and through release of PGE2 [102]. In accordance with their role as
modulators of the immune response in tumors, depletion of CAFs from experimental
breast cancers – via DNA vaccination against fibroblast activation protein (FAP) – results
in suppression of metastatic spread through shifting the polarization of the immnune
microenvironment from Th2 to Th1 [103].
6. The CAF as a Clinical Entity
22
Given the importance of stromal support in carcinogenesis and tumor progression, CAFs
are thought to provide an important target for therapeutic strategies to improve outcomes
for cancer patients. In the normal breast, fibroblasts are quite abundant and, in the case
of carcinogenesis, infiltration of CAFs and their ECM are the reason why breast cancer
often presents as a palpable lump. Additionally, high mammographic density is
associated with an increased risk of developing breast cancer [104]. Recent clinical and
translational studies have made it clear that the extent of stromal desmoplasia is linked to
prognosis and that molecular profiling of stromal markers is predictive of outcome [105,
106]. In the clinic, a myofibroblast-rich, reactive stroma is almost always found in human
adenocarcinomas and is associated with invasiveness and poor prognosis [20, 107].
Understanding the heterogeneity within fibroblast populations will be important if
we are to effectively target the tumor promoting functions of these cells. Indeed,
fibroblasts express differential levels of key interleukins and chemokines in breast
cancers of different subtypes (i.e., basal-like and luminal-like) [108]. An interesting
recent study provided a plausible explanation for why triple negative breast cancer (ER-
/PR-/HER2-; TNBC) is more prevalent among african american women than caucasion
women by showing that normal mammary fibroblasts from afterican american women
support the growth of TNBC cells while those from caucasian women did not [109]. As
TNBC denotes poor prognosis and there are not targeted therapies against this kind of
tumor, studies aiming to elucidate why certain fibroblasts promote one kind of disease or
another are crucial.
Unfortunately, however, there are no current therapies deisgned to directly inhibit
or eliminate tumor-promoting CAFs. Currently, an ongoing clinical trial is investigating
23
the effect of irradiation of the area around non-invasive or early invasive breast cancer
(http://www.clinicaltrials.gov). This trial is based on pre-clinical data that support the
role of the microenvironment in progression and invasion of breast cancer; however, in
some pre-clinical models irradiation of the mammary gland actually promotes
transformation of immortalized tumor cells [110]. A separate study on pancreatic tumors
suggested that blocking HGF signaling in addition to irradiation of the stroma may help
prevent invasion of cancerous epithelial cells [111].
Perhaps indirect targeting of other components in the tumor microenvironment
will serve to inhibit CAF accumulation or function. For example, certain tyrosine kinase
inhibitors that target signaling pathways (e.g., VEGFR, PDGFR) that are active in
stromal cell populations (e.g. endothelial cells and fibroblasts) have been proposed. In the
past decades, there has been much interest in the engineering of anti-angiogenesis
molecules that prevent the angiogenic switch in early malignancies and thereby prevent
tumor growth. Unfortunately, these efforts have been met with disappointment, as
patients often acquire resistance to drugs like Avastin (VEGFA antibody), which has
recently been revoked from clinical trials for breast cancer [112]. Studies show that
cancer cells can evolve resistance to blockade of the VEGF pathway; however,
concomitant inhibition of FGF and VEGF during this process results in tumor stasis and a
reduction in tumor burden [113]. Conversely, results from pre-clinical models have lead
to suggestions that enhancing tumor vascularization (e.g., inhibiton of PDGF-B or
hedgehog) might help improve delivery of chemotherapy, thereby attenuating tumor
progression [94, 114].
24
Targeting the structure and homeostasis of the abundant ECM found within
tumors may one day prove useful in therapeutic approaches as well. For example, lysyl
oxidase enzymes, responsible for collagen cross-linking, are elevated in many cancers
[115]; however, the ECM turnover in these conditions is abnormal. Although strategies to
inhibit MMP activity have been proposed [116], clinical trials using MMP inhibitors
yielded disappointing results due to serious toxicity and lack of specificity [117].
Finally, targeting factors that promote and sustain the CAF phenotype, as
demonstrated in pre-clinical models, might lead to novel therapeutic approaches. For
example, targeting BMDC-derived granulin, which was shown to induce the tumor-
promoting function of normal mammary fibroblasts, might provide a means to prevent
CAF support of cancer progression [19]. Autocrine factors, such as TGFb and SDF-1,
that drive and sustain CAF populations might provide a basis for exploring ways to
inhibit the tumor support network [118]. Likewise, elucidating some of the currently
unknown factors derived from cancer cells that promote CAF function [18, 119] should
also lead to identification of therapeutic candidates.
7. Perspectives
The aims of many pre-clinical cancer research efforts have been to elucidate mechanisms
by which certain tumors progress toward life-threatening malignancy and acquire
resistance to treatments. As such, many studies consider the variety of cell types that
comprise a solid tumor. In that vain, if targeted therapies are to be successful, then one
must take into account not only the molecular target of a given therapy, but also the
25
patient population that is most likely to respond to that therapy. Also, delineating
pathways that provide tumors with adaptive resistence to current therapies allows for the
opportunity to attack tumors via multiple modalities. It is likely that understanding
similarities between cancer pathogenesis, fibrosis, and wound healing will be a step in
that direction. Defining CAFs, understanding their behavior, and elucidating how they
arise is one part of this challenging task and holds the promise of desigining ways to
attack tumor-supportive CAFs, perhaps while sparing normal fibroblasts.
Acknowledgements
We thank Dr. Brian Bierie for helpful discussion. KP is the Göran and Birgitta Grosskopf
Professor of Molecular Medicine at Lund University, and is supported by a Linnaeus
grant to the STARGET consortium and by the ERC Starting Grant TUMORGAN. SSM
is supported by the Harvard Stem Cell Institute, the Brigham Research Institute, and NIH
NCI RO1 CA166284-01.
References
[1] M. Jacob, L. Chang, E. Pure, Fibroblast Activation Protein in Remodeling Tissues, Current molecular medicine, (2012). [2] R. Kalluri, Basement membranes: structure, assembly and role in tumour angiogenesis, Nature reviews. Cancer, 3 (2003) 422-‐433. [3] L. Ronnov-‐Jessen, O.W. Petersen, M.J. Bissell, Cellular changes involved in conversion of normal to malignant breast: importance of the stromal reaction, Physiological reviews, 76 (1996) 69-‐125. [4] A.T. Nurden, Platelets, inflammation and tissue regeneration, Thrombosis and haemostasis, 105 Suppl 1 (2011) S13-‐33. [5] H. Wiig, K. Rubin, R.K. Reed, New and active role of the interstitium in control of interstitial fluid pressure: potential therapeutic consequences, Acta anaesthesiologica Scandinavica, 47 (2003) 111-‐121. [6] H.F. Dvorak, Tumors: wounds that do not heal. Similarities between tumor stroma generation and wound healing, The New England journal of medicine, 315 (1986) 1650-‐1659.
26
[7] A.F. Olumi, G.D. Grossfeld, S.W. Hayward, P.R. Carroll, T.D. Tlsty, G.R. Cunha, Carcinoma-‐associated fibroblasts direct tumor progression of initiated human prostatic epithelium, Cancer research, 59 (1999) 5002-‐5011. [8] M. Hu, J. Yao, D.K. Carroll, S. Weremowicz, H. Chen, D. Carrasco, A. Richardson, S. Violette, T. Nikolskaya, Y. Nikolsky, E.L. Bauerlein, W.C. Hahn, R.S. Gelman, C. Allred, M.J. Bissell, S. Schnitt, K. Polyak, Regulation of in situ to invasive breast carcinoma transition, Cancer cell, 13 (2008) 394-‐406. [9] R. Hill, Y. Song, R.D. Cardiff, T. Van Dyke, Selective evolution of stromal mesenchyme with p53 loss in response to epithelial tumorigenesis, Cell, 123 (2005) 1001-‐1011. [10] N.A. Bhowmick, A. Chytil, D. Plieth, A.E. Gorska, N. Dumont, S. Shappell, M.K. Washington, E.G. Neilson, H.L. Moses, TGF-‐beta signaling in fibroblasts modulates the oncogenic potential of adjacent epithelia, Science, 303 (2004) 848-‐851. [11] C. Anderberg, H. Li, L. Fredriksson, J. Andrae, C. Betsholtz, X. Li, U. Eriksson, K. Pietras, Paracrine signaling by platelet-‐derived growth factor-‐CC promotes tumor growth by recruitment of cancer-‐associated fibroblasts, Cancer research, 69 (2009) 369-‐378. [12] P. Micke, A. Ostman, Tumour-‐stroma interaction: cancer-‐associated fibroblasts as novel targets in anti-‐cancer therapy?, Lung cancer, 45 Suppl 2 (2004) S163-‐175. [13] H. Sugimoto, T.M. Mundel, M.W. Kieran, R. Kalluri, Identification of fibroblast heterogeneity in the tumor microenvironment, Cancer biology & therapy, 5 (2006) 1640-‐1646. [14] J. Paulsson, T. Sjoblom, P. Micke, F. Ponten, G. Landberg, C.H. Heldin, J. Bergh, D.J. Brennan, K. Jirstrom, A. Ostman, Prognostic significance of stromal platelet-‐derived growth factor beta-‐receptor expression in human breast cancer, The American journal of pathology, 175 (2009) 334-‐341. [15] A. Silini, C. Ghilardi, S. Figini, F. Sangalli, R. Fruscio, R. Dahse, R.B. Pedley, R. Giavazzi, M. Bani, Regulator of G-‐protein signaling 5 (RGS5) protein: a novel marker of cancer vasculature elicited and sustained by the tumor's proangiogenic microenvironment, Cellular and molecular life sciences : CMLS, 69 (2012) 1167-‐1178. [16] J. Tchou, A.V. Kossenkov, L. Chang, C. Satija, M. Herlyn, L.C. Showe, E. Pure, Human breast cancer associated fibroblasts exhibit subtype specific gene expression profiles, BMC medical genomics, 5 (2012) 39. [17] J.A. Rudnick, L.M. Arendt, I. Klebba, J.W. Hinds, V. Iyer, P.B. Gupta, S.P. Naber, C. Kuperwasser, Functional heterogeneity of breast fibroblasts is defined by a prostaglandin secretory phenotype that promotes expansion of cancer-‐stem like cells, PLoS One, 6 (2011) e24605. [18] N. Erez, M. Truitt, P. Olson, S.T. Arron, D. Hanahan, Cancer-‐Associated Fibroblasts Are Activated in Incipient Neoplasia to Orchestrate Tumor-‐Promoting Inflammation in an NF-‐kappaB-‐Dependent Manner, Cancer cell, 17 (2010) 135-‐147. [19] M. Elkabets, A.M. Gifford, C. Scheel, B. Nilsson, F. Reinhardt, M.A. Bray, A.E. Carpenter, K. Jirstrom, K. Magnusson, B.L. Ebert, F. Ponten, R.A. Weinberg, S.S. McAllister, Human tumors instigate granulin-‐expressing hematopoietic cells that promote malignancy by activating stromal fibroblasts in mice, The Journal of clinical investigation, 121 (2011) 784-‐799. [20] R. Kalluri, M. Zeisberg, Fibroblasts in cancer, Nature reviews. Cancer, 6 (2006) 392-‐401. [21] A.J. Trimboli, C.Z. Cantemir-‐Stone, F. Li, J.A. Wallace, A. Merchant, N. Creasap, J.C. Thompson, E. Caserta, H. Wang, J.L. Chong, S. Naidu, G. Wei, S.M. Sharma, J.A. Stephens, S.A. Fernandez, M.N. Gurcan, M.B. Weinstein, S.H. Barsky, L. Yee, T.J. Rosol, P.C. Stromberg, M.L. Robinson, F. Pepin, M. Hallett, M. Park, M.C. Ostrowski, G. Leone, Pten in stromal fibroblasts suppresses mammary epithelial tumours, Nature, 461 (2009) 1084-‐1091.
27
[22] S.W. Hayward, Y. Wang, M. Cao, Y.K. Hom, B. Zhang, G.D. Grossfeld, D. Sudilovsky, G.R. Cunha, Malignant transformation in a nontumorigenic human prostatic epithelial cell line, Cancer research, 61 (2001) 8135-‐8142. [23] S.F. Jue, R.S. Bradley, J.A. Rudnicki, H.E. Varmus, A.M. Brown, The mouse Wnt-‐1 gene can act via a paracrine mechanism in transformation of mammary epithelial cells, Molecular and cellular biology, 12 (1992) 321-‐328. [24] C. Kuperwasser, T. Chavarria, M. Wu, G. Magrane, J.W. Gray, L. Carey, A. Richardson, R.A. Weinberg, Reconstruction of functionally normal and malignant human breast tissues in mice, Proceedings of the National Academy of Sciences of the United States of America, 101 (2004) 4966-‐4971. [25] B. Grum-‐Schwensen, J. Klingelhofer, C.H. Berg, C. El-‐Naaman, M. Grigorian, E. Lukanidin, N. Ambartsumian, Suppression of tumor development and metastasis formation in mice lacking the S100A4(mts1) gene, Cancer research, 65 (2005) 3772-‐3780. [26] R. Kalluri, E.G. Neilson, Epithelial-‐mesenchymal transition and its implications for fibrosis, J Clin Invest, 112 (2003) 1776-‐1784. [27] M.A. Cichon, A.C. Degnim, D.W. Visscher, D.C. Radisky, Microenvironmental influences that drive progression from benign breast disease to invasive breast cancer, Journal of mammary gland biology and neoplasia, 15 (2010) 389-‐397. [28] D.H. Nguyen, H.A. Oketch-‐Rabah, I. Illa-‐Bochaca, F.C. Geyer, J.S. Reis-‐Filho, J.H. Mao, S.A. Ravani, J. Zavadil, A.D. Borowsky, D.J. Jerry, K.A. Dunphy, J.H. Seo, S. Haslam, D. Medina, M.H. Barcellos-‐Hoff, Radiation acts on the microenvironment to affect breast carcinogenesis by distinct mechanisms that decrease cancer latency and affect tumor type, Cancer cell, 19 (2011) 640-‐651. [29] E. Pazolli, X. Luo, S. Brehm, K. Carbery, J.J. Chung, J.L. Prior, J. Doherty, S. Demehri, L. Salavaggione, D. Piwnica-‐Worms, S.A. Stewart, Senescent stromal-‐derived osteopontin promotes preneoplastic cell growth, Cancer Res, 69 (2009) 1230-‐1239. [30] K.R. Levental, H. Yu, L. Kass, J.N. Lakins, M. Egeblad, J.T. Erler, S.F. Fong, K. Csiszar, A. Giaccia, W. Weninger, M. Yamauchi, D.L. Gasser, V.M. Weaver, Matrix crosslinking forces tumor progression by enhancing integrin signaling, Cell, 139 (2009) 891-‐906. [31] F. Blasi, N. Sidenius, The urokinase receptor: focused cell surface proteolysis, cell adhesion and signaling, FEBS letters, 584 (2010) 1923-‐1930. [32] B. Cat, D. Stuhlmann, H. Steinbrenner, L. Alili, O. Holtkotter, H. Sies, P. Brenneisen, Enhancement of tumor invasion depends on transdifferentiation of skin fibroblasts mediated by reactive oxygen species, Journal of cell science, 119 (2006) 2727-‐2738. [33] A. Toullec, D. Gerald, G. Despouy, B. Bourachot, M. Cardon, S. Lefort, M. Richardson, G. Rigaill, M.C. Parrini, C. Lucchesi, D. Bellanger, M.H. Stern, T. Dubois, X. Sastre-‐Garau, O. Delattre, A. Vincent-‐Salomon, F. Mechta-‐Grigoriou, Oxidative stress promotes myofibroblast differentiation and tumour spreading, EMBO molecular medicine, 2 (2010) 211-‐230. [34] U.E. Martinez-‐Outschoorn, C. Trimmer, Z. Lin, D. Whitaker-‐Menezes, B. Chiavarina, J. Zhou, C. Wang, S. Pavlides, M.P. Martinez-‐Cantarin, F. Capozza, A.K. Witkiewicz, N. Flomenberg, A. Howell, R.G. Pestell, J. Caro, M.P. Lisanti, F. Sotgia, Autophagy in cancer associated fibroblasts promotes tumor cell survival: Role of hypoxia, HIF1 induction and NFkappaB activation in the tumor stromal microenvironment, Cell cycle, 9 (2010) 3515-‐3533. [35] R. Kalluri, R.A. Weinberg, The basics of epithelial-‐mesenchymal transition, The Journal of clinical investigation, 119 (2009) 1420-‐1428. [36] S.A. Mani, W. Guo, M.J. Liao, E.N. Eaton, A. Ayyanan, A.Y. Zhou, M. Brooks, F. Reinhard, C.C. Zhang, M. Shipitsin, L.L. Campbell, K. Polyak, C. Brisken, J. Yang, R.A. Weinberg, The epithelial-‐mesenchymal transition generates cells with properties of stem cells, Cell, 133 (2008) 704-‐715.
28
[37] W.B. Fang, I. Jokar, A. Chytil, H.L. Moses, T. Abel, N. Cheng, Loss of one Tgfbr2 allele in fibroblasts promotes metastasis in MMTV: polyoma middle T transgenic and transplant mouse models of mammary tumor progression, Clinical & experimental metastasis, 28 (2011) 351-‐366. [38] C. Gaggioli, S. Hooper, C. Hidalgo-‐Carcedo, R. Grosse, J.F. Marshall, K. Harrington, E. Sahai, Fibroblast-‐led collective invasion of carcinoma cells with differing roles for RhoGTPases in leading and following cells, Nature cell biology, 9 (2007) 1392-‐1400. [39] J.G. Goetz, S. Minguet, I. Navarro-‐Lerida, J.J. Lazcano, R. Samaniego, E. Calvo, M. Tello, T. Osteso-‐Ibanez, T. Pellinen, A. Echarri, A. Cerezo, A.J. Klein-‐Szanto, R. Garcia, P.J. Keely, P. Sanchez-‐Mateos, E. Cukierman, M.A. Del Pozo, Biomechanical remodeling of the microenvironment by stromal caveolin-‐1 favors tumor invasion and metastasis, Cell, 146 (2011) 148-‐163. [40] E. Olaso, C. Salado, E. Egilegor, V. Gutierrez, A. Santisteban, P. Sancho-‐Bru, S.L. Friedman, F. Vidal-‐Vanaclocha, Proangiogenic role of tumor-‐activated hepatic stellate cells in experimental melanoma metastasis, Hepatology, 37 (2003) 674-‐685. [41] I. Malanchi, A. Santamaria-‐Martinez, E. Susanto, H. Peng, H.A. Lehr, J.F. Delaloye, J. Huelsken, Interactions between cancer stem cells and their niche govern metastatic colonization, Nature, 481 (2012) 85-‐89. [42] H.S. Kuznetsov, T. Marsh, B.A. Markens, Z. Castano, A. Greene-‐Colozzi, S.A. Hay, V.E. Brown, A.L. Richardson, S. Signoretti, E.M. Battinelli, S.S. McAllister, Identification of Luminal Breast Cancers that Establish a Tumor Supportive Macroenvironment Defined by Pro-‐Angiogenic Platelets and Bone Marrow Derived Cells, Cancer discovery, (2012). [43] D.G. Duda, A.M. Duyverman, M. Kohno, M. Snuderl, E.J. Steller, D. Fukumura, R.K. Jain, Malignant cells facilitate lung metastasis by bringing their own soil, Proceedings of the National Academy of Sciences of the United States of America, 107 (2010) 21677-‐21682. [44] M. Zhao, P.C. Sachs, X. Wang, C.I. Dumur, M.O. Idowu, V. Robila, M.P. Francis, J. Ware, M. Beckman, A. Rizki, S.E. Holt, L.W. Elmore, Mesenchymal stem cells in mammary adipose tissue stimulate progression of breast cancer resembling the basal-‐type, Cancer biology & therapy, 13 (2012). [45] N. Walker, L. Badri, S. Wettlaufer, A. Flint, U. Sajjan, P.H. Krebsbach, V.G. Keshamouni, M. Peters-‐Golden, V.N. Lama, Resident tissue-‐specific mesenchymal progenitor cells contribute to fibrogenesis in human lung allografts, The American journal of pathology, 178 (2011) 2461-‐2469. [46] M. Razmkhah, M. Jaberipour, A. Hosseini, A. Safaei, B. Khalatbari, A. Ghaderi, Expression profile of IL-‐8 and growth factors in breast cancer cells and adipose-‐derived stem cells (ASCs) isolated from breast carcinoma, Cellular immunology, 265 (2010) 80-‐85. [47] M. Quante, S.P. Tu, H. Tomita, T. Gonda, S.S. Wang, S. Takashi, G.H. Baik, W. Shibata, B. Diprete, K.S. Betz, R. Friedman, A. Varro, B. Tycko, T.C. Wang, Bone marrow-‐derived myofibroblasts contribute to the mesenchymal stem cell niche and promote tumor growth, Cancer cell, 19 (2011) 257-‐272. [48] G. Ishii, T. Sangai, T. Oda, Y. Aoyagi, T. Hasebe, N. Kanomata, Y. Endoh, C. Okumura, Y. Okuhara, J. Magae, M. Emura, T. Ochiya, A. Ochiai, Bone-‐marrow-‐derived myofibroblasts contribute to the cancer-‐induced stromal reaction, Biochem Biophys Res Commun, 309 (2003) 232-‐240. [49] G. Ishii, T. Sangai, T. Ito, T. Hasebe, Y. Endoh, H. Sasaki, K. Harigaya, A. Ochiai, In vivo and in vitro characterization of human fibroblasts recruited selectively into human cancer stroma, Int J Cancer, 117 (2005) 212-‐220. [50] N.C. Direkze, K. Hodivala-‐Dilke, R. Jeffery, T. Hunt, R. Poulsom, D. Oukrif, M.R. Alison, N.A. Wright, Bone marrow contribution to tumor-‐associated myofibroblasts and fibroblasts, Cancer Res, 64 (2004) 8492-‐8495.
29
[51] N.C. Direkze, M.R. Alison, Bone marrow and tumour stroma: an intimate relationship, Hematol Oncol, (2006). [52] B. Hall, M. Andreeff, F. Marini, The participation of mesenchymal stem cells in tumor stroma formation and their application as targeted-‐gene delivery vehicles, Handbook of experimental pharmacology, (2007) 263-‐283. [53] S.C. Hung, W.P. Deng, W.K. Yang, R.S. Liu, C.C. Lee, T.C. Su, R.J. Lin, D.M. Yang, C.W. Chang, W.H. Chen, H.J. Wei, J.G. Gelovani, Mesenchymal stem cell targeting of microscopic tumors and tumor stroma development monitored by noninvasive in vivo positron emission tomography imaging, Clinical cancer research : an official journal of the American Association for Cancer Research, 11 (2005) 7749-‐7756. [54] S. Kidd, E. Spaeth, J.L. Dembinski, M. Dietrich, K. Watson, A. Klopp, V.L. Battula, M. Weil, M. Andreeff, F.C. Marini, Direct evidence of mesenchymal stem cell tropism for tumor and wounding microenvironments using in vivo bioluminescent imaging, Stem Cells, 27 (2009) 2614-‐2623. [55] R.M. Dwyer, S.M. Potter-‐Beirne, K.A. Harrington, A.J. Lowery, E. Hennessy, J.M. Murphy, F.P. Barry, T. O'Brien, M.J. Kerin, Monocyte chemotactic protein-‐1 secreted by primary breast tumors stimulates migration of mesenchymal stem cells, Clinical cancer research : an official journal of the American Association for Cancer Research, 13 (2007) 5020-‐5027. [56] B. Feng, L. Chen, Review of mesenchymal stem cells and tumors: executioner or coconspirator?, Cancer biotherapy & radiopharmaceuticals, 24 (2009) 717-‐721. [57] E.L. Spaeth, J.L. Dembinski, A.K. Sasser, K. Watson, A. Klopp, B. Hall, M. Andreeff, F. Marini, Mesenchymal stem cell transition to tumor-‐associated fibroblasts contributes to fibrovascular network expansion and tumor progression, PLoS One, 4 (2009) e4992. [58] A.E. Karnoub, A.B. Dash, A.P. Vo, A. Sullivan, M.W. Brooks, G.W. Bell, A.L. Richardson, K. Polyak, R. Tubo, R.A. Weinberg, Mesenchymal stem cells within tumour stroma promote breast cancer metastasis, Nature, 449 (2007) 557-‐563. [59] R. Abe, S.C. Donnelly, T. Peng, R. Bucala, C.N. Metz, Peripheral blood fibrocytes: differentiation pathway and migration to wound sites, Journal of immunology, 166 (2001) 7556-‐7562. [60] R. Bucala, L.A. Spiegel, J. Chesney, M. Hogan, A. Cerami, Circulating fibrocytes define a new leukocyte subpopulation that mediates tissue repair, Molecular medicine, 1 (1994) 71-‐81. [61] M. Schmidt, G. Sun, M.A. Stacey, L. Mori, S. Mattoli, Identification of circulating fibrocytes as precursors of bronchial myofibroblasts in asthma, Journal of immunology, 171 (2003) 380-‐389. [62] N. Hashimoto, H. Jin, T. Liu, S.W. Chensue, S.H. Phan, Bone marrow-‐derived progenitor cells in pulmonary fibrosis, The Journal of clinical investigation, 113 (2004) 243-‐252. [63] E.C. Keeley, B. Mehrad, R.M. Strieter, The role of fibrocytes in fibrotic diseases of the lungs and heart, Fibrogenesis & tissue repair, 4 (2011) 2. [64] B.D. Humphreys, S.L. Lin, A. Kobayashi, T.E. Hudson, B.T. Nowlin, J.V. Bonventre, M.T. Valerius, A.P. McMahon, J.S. Duffield, Fate tracing reveals the pericyte and not epithelial origin of myofibroblasts in kidney fibrosis, The American journal of pathology, 176 (2010) 85-‐97. [65] S. Dulauroy, S.E. Di Carlo, F. Langa, G. Eberl, L. Peduto, Lineage tracing and genetic ablation of ADAM12(+) perivascular cells identify a major source of profibrotic cells during acute tissue injury, Nature medicine, (2012). [66] C. Goritz, D.O. Dias, N. Tomilin, M. Barbacid, O. Shupliakov, J. Frisen, A pericyte origin of spinal cord scar tissue, Science, 333 (2011) 238-‐242. [67] H.Y. Chang, J.B. Sneddon, A.A. Alizadeh, R. Sood, R.B. West, K. Montgomery, J.T. Chi, M. van de Rijn, D. Botstein, P.O. Brown, Gene expression signature of fibroblast serum response
30
predicts human cancer progression: similarities between tumors and wounds, PLoS Biol, 2 (2004) E7. [68] D.M. McDonald, P.L. Choyke, Imaging of angiogenesis: from microscope to clinic, Nature medicine, 9 (2003) 713-‐725. [69] S. Kidd, E. Spaeth, K. Watson, J. Burks, H. Lu, A. Klopp, M. Andreeff, F.C. Marini, Origins of the tumor microenvironment: quantitative assessment of adipose-‐derived and bone marrow-‐derived stroma, PloS one, 7 (2012) e30563. [70] E. Devarajan, Y.H. Song, S. Krishnappa, E. Alt, Epithelial-‐mesenchymal transition in breast cancer lines is mediated through PDGF-‐D released by tissue-‐resident stem cells, International journal of cancer. Journal international du cancer, 131 (2012) 1023-‐1031. [71] L. Meng, J. Zhou, H. Sasano, T. Suzuki, K.M. Zeitoun, S.E. Bulun, Tumor necrosis factor alpha and interleukin 11 secreted by malignant breast epithelial cells inhibit adipocyte differentiation by selectively down-‐regulating CCAAT/enhancer binding protein alpha and peroxisome proliferator-‐activated receptor gamma: mechanism of desmoplastic reaction, Cancer research, 61 (2001) 2250-‐2255. [72] E.M. Zeisberg, S.E. Potenta, H. Sugimoto, M. Zeisberg, R. Kalluri, Fibroblasts in kidney fibrosis emerge via endothelial-‐to-‐mesenchymal transition, J Am Soc Nephrol, 19 (2008) 2282-‐2287. [73] D. Medici, R. Kalluri, Endothelial-‐mesenchymal transition and its contribution to the emergence of stem cell phenotype, Seminars in cancer biology, (2012). [74] E.M. Zeisberg, S. Potenta, L. Xie, M. Zeisberg, R. Kalluri, Discovery of endothelial to mesenchymal transition as a source for carcinoma-‐associated fibroblasts, Cancer research, 67 (2007) 10123-‐10128. [75] J. Yang, R.A. Weinberg, Epithelial-‐mesenchymal transition: at the crossroads of development and tumor metastasis, Developmental cell, 14 (2008) 818-‐829. [76] D.C. Radisky, P.A. Kenny, M.J. Bissell, Fibrosis and cancer: do myofibroblasts come also from epithelial cells via EMT?, Journal of cellular biochemistry, 101 (2007) 830-‐839. [77] C.J. Creighton, J.C. Chang, J.M. Rosen, Epithelial-‐mesenchymal transition (EMT) in tumor-‐initiating cells and its clinical implications in breast cancer, Journal of mammary gland biology and neoplasia, 15 (2010) 253-‐260. [78] Y. Wang, B.P. Zhou, Epithelial-‐mesenchymal transition in breast cancer progression and metastasis, Chinese journal of cancer, 30 (2011) 603-‐611. [79] O.W. Petersen, H.L. Nielsen, T. Gudjonsson, R. Villadsen, F. Rank, E. Niebuhr, M.J. Bissell, L. Ronnov-‐Jessen, Epithelial to mesenchymal transition in human breast cancer can provide a nonmalignant stroma, The American journal of pathology, 162 (2003) 391-‐402. [80] D. Granot, Y. Addadi, V. Kalchenko, A. Harmelin, L.A. Kunz-‐Schughart, M. Neeman, In vivo imaging of the systemic recruitment of fibroblasts to the angiogenic rim of ovarian carcinoma tumors, Cancer Res, 67 (2007) 9180-‐9189. [81] D. Fukumura, R. Xavier, T. Sugiura, Y. Chen, E.C. Park, N. Lu, M. Selig, G. Nielsen, T. Taksir, R.K. Jain, B. Seed, Tumor induction of VEGF promoter activity in stromal cells, Cell, 94 (1998) 715-‐725. [82] V.L. Thijssen, R.J. Brandwijk, R.P. Dings, A.W. Griffioen, Angiogenesis gene expression profiling in xenograft models to study cellular interactions, Exp Cell Res, 299 (2004) 286-‐293. [83] A.A. Gilad, T. Israely, H. Dafni, G. Meir, B. Cohen, M. Neeman, Functional and molecular mapping of uncoupling between vascular permeability and loss of vascular maturation in ovarian carcinoma xenografts: the role of stroma cells in tumor angiogenesis, Int J Cancer, 117 (2005) 202-‐211.
31
[84] K. Pietras, A. Ostman, Hallmarks of cancer: interactions with the tumor stroma, Exp Cell Res, 316 (2010) 1324-‐1331. [85] K. Pietras, J. Pahler, G. Bergers, D. Hanahan, Functions of paracrine PDGF signaling in the proangiogenic tumor stroma revealed by pharmacological targeting, PLoS Med, 5 (2008) e19. [86] D. Leali, P. Dell'Era, H. Stabile, B. Sennino, A.F. Chambers, A. Naldini, S. Sozzani, B. Nico, D. Ribatti, M. Presta, Osteopontin (Eta-‐1) and fibroblast growth factor-‐2 cross-‐talk in angiogenesis, J Immunol, 171 (2003) 1085-‐1093. [87] J. Dai, L. Peng, K. Fan, H. Wang, R. Wei, G. Ji, J. Cai, B. Lu, B. Li, D. Zhang, Y. Kang, M. Tan, W. Qian, Y. Guo, Osteopontin induces angiogenesis through activation of PI3K/AKT and ERK1/2 in endothelial cells, Oncogene, 28 (2009) 3412-‐3422. [88] G. Chakraborty, S. Jain, G.C. Kundu, Osteopontin promotes vascular endothelial growth factor-‐dependent breast tumor growth and angiogenesis via autocrine and paracrine mechanisms, Cancer Res, 68 (2008) 152-‐161. [89] J. Dong, J. Grunstein, M. Tejada, F. Peale, G. Frantz, W.C. Liang, W. Bai, L. Yu, J. Kowalski, X. Liang, G. Fuh, H.P. Gerber, N. Ferrara, VEGF-‐null cells require PDGFR alpha signaling-‐mediated stromal fibroblast recruitment for tumorigenesis, Embo J, 23 (2004) 2800-‐2810. [90] Y. Crawford, I. Kasman, L. Yu, C. Zhong, X. Wu, Z. Modrusan, J. Kaminker, N. Ferrara, PDGF-‐C mediates the angiogenic and tumorigenic properties of fibroblasts associated with tumors refractory to anti-‐VEGF treatment, Cancer Cell, 15 (2009) 21-‐34. [91] M. Augsten, C. Hagglof, E. Olsson, C. Stolz, P. Tsagozis, T. Levchenko, M.J. Frederick, A. Borg, P. Micke, L. Egevad, A. Ostman, CXCL14 is an autocrine growth factor for fibroblasts and acts as a multi-‐modal stimulator of prostate tumor growth, Proc Natl Acad Sci U S A, 106 (2009) 3414-‐3419. [92] S. Vong, R. Kalluri, The role of stromal myofibroblast and extracellular matrix in tumor angiogenesis, Genes Cancer, 2 (2012) 1139-‐1145. [93] R.L. Yauch, S.E. Gould, S.J. Scales, T. Tang, H. Tian, C.P. Ahn, D. Marshall, L. Fu, T. Januario, D. Kallop, M. Nannini-‐Pepe, K. Kotkow, J.C. Marsters, L.L. Rubin, F.J. de Sauvage, A paracrine requirement for hedgehog signalling in cancer, Nature, 455 (2008) 406-‐410. [94] K.P. Olive, M.A. Jacobetz, C.J. Davidson, A. Gopinathan, D. McIntyre, D. Honess, B. Madhu, M.A. Goldgraben, M.E. Caldwell, D. Allard, K.K. Frese, G. Denicola, C. Feig, C. Combs, S.P. Winter, H. Ireland-‐Zecchini, S. Reichelt, W.J. Howat, A. Chang, M. Dhara, L. Wang, F. Ruckert, R. Grutzmann, C. Pilarsky, K. Izeradjene, S.R. Hingorani, P. Huang, S.E. Davies, W. Plunkett, M. Egorin, R.H. Hruban, N. Whitebread, K. McGovern, J. Adams, C. Iacobuzio-‐Donahue, J. Griffiths, D.A. Tuveson, Inhibition of Hedgehog signaling enhances delivery of chemotherapy in a mouse model of pancreatic cancer, Science, 324 (2009) 1457-‐1461. [95] A. Orimo, P.B. Gupta, D.C. Sgroi, F. Arenzana-‐Seisdedos, T. Delaunay, R. Naeem, V.J. Carey, A.L. Richardson, R.A. Weinberg, Stromal fibroblasts present in invasive human breast carcinomas promote tumor growth and angiogenesis through elevated SDF-‐1/CXCL12 secretion, Cell, 121 (2005) 335-‐348. [96] D. Hanahan, R.A. Weinberg, Hallmarks of cancer: the next generation, Cell, 144 (2011) 646-‐674. [97] N. Erez, M. Truitt, P. Olson, D. Hanahan, Cancer-‐Associated Fibroblasts Are Activated in Incipient Neoplasia to Orchestrate Tumor-‐Promoting Inflammation in an NF-‐kappaB-‐Dependent Manner, Cancer Cell, 17 (2010) 135-‐147. [98] T. Silzle, M. Kreutz, M.A. Dobler, G. Brockhoff, R. Knuechel, L.A. Kunz-‐Schughart, Tumor-‐associated fibroblasts recruit blood monocytes into tumor tissue, Eur J Immunol, 33 (2003) 1311-‐1320.
32
[99] B.J. Boersma, M. Reimers, M. Yi, J.A. Ludwig, B.T. Luke, R.M. Stephens, H.G. Yfantis, D.H. Lee, J.N. Weinstein, S. Ambs, A stromal gene signature associated with inflammatory breast cancer, Int J Cancer, 122 (2008) 1324-‐1332. [100] B. Grum-‐Schwensen, J. Klingelhofer, M. Grigorian, K. Almholt, B.S. Nielsen, E. Lukanidin, N. Ambartsumian, Lung metastasis fails in MMTV-‐PyMT oncomice lacking S100A4 due to a T-‐cell deficiency in primary tumors, Cancer Res, 70 (2010) 936-‐947. [101] D.G. Stover, B. Bierie, H.L. Moses, A delicate balance: TGF-‐beta and the tumor microenvironment, J Cell Biochem, 101 (2007) 851-‐861. [102] M. Balsamo, F. Scordamaglia, G. Pietra, C. Manzini, C. Cantoni, M. Boitano, P. Queirolo, W. Vermi, F. Facchetti, A. Moretta, L. Moretta, M.C. Mingari, M. Vitale, Melanoma-‐associated fibroblasts modulate NK cell phenotype and antitumor cytotoxicity, Proc Natl Acad Sci U S A, 106 (2009) 20847-‐20852. [103] D. Liao, Y. Luo, D. Markowitz, R. Xiang, R.A. Reisfeld, Cancer associated fibroblasts promote tumor growth and metastasis by modulating the tumor immune microenvironment in a 4T1 murine breast cancer model, PLoS One, 4 (2009) e7965. [104] V. Assi, J. Warwick, J. Cuzick, S.W. Duffy, Clinical and epidemiological issues in mammographic density, Nature reviews. Clinical oncology, 9 (2012) 33-‐40. [105] G. Finak, N. Bertos, F. Pepin, S. Sadekova, M. Souleimanova, H. Zhao, H. Chen, G. Omeroglu, S. Meterissian, A. Omeroglu, M. Hallett, M. Park, Stromal gene expression predicts clinical outcome in breast cancer, Nature medicine, 14 (2008) 518-‐527. [106] A. Planche, M. Bacac, P. Provero, C. Fusco, M. Delorenzi, J.C. Stehle, I. Stamenkovic, Identification of prognostic molecular features in the reactive stroma of human breast and prostate cancer, PloS one, 6 (2011) e18640. [107] R.A. Walker, Are all ductal proliferations of the breast premalignant?, The Journal of pathology, 195 (2001) 401-‐403. [108] J.T. Camp, F. Elloumi, E. Roman-‐Perez, J. Rein, D.A. Stewart, J.C. Harrell, C.M. Perou, M.A. Troester, Interactions with fibroblasts are distinct in Basal-‐like and luminal breast cancers, Molecular cancer research : MCR, 9 (2011) 3-‐13. [109] J.M. Fleming, T.C. Miller, M. Quinones, Z. Xiao, X. Xu, M.J. Meyer, E. Ginsburg, T.D. Veenstra, B.K. Vonderhaar, The normal breast microenvironment of premenopausal women differentially influences the behavior of breast cancer cells in vitro and in vivo, BMC medicine, 8 (2010) 27. [110] M.H. Barcellos-‐Hoff, S.A. Ravani, Irradiated mammary gland stroma promotes the expression of tumorigenic potential by unirradiated epithelial cells, Cancer research, 60 (2000) 1254-‐1260. [111] K. Ohuchida, K. Mizumoto, M. Murakami, L.W. Qian, N. Sato, E. Nagai, K. Matsumoto, T. Nakamura, M. Tanaka, Radiation to stromal fibroblasts increases invasiveness of pancreatic cancer cells through tumor-‐stromal interactions, Cancer research, 64 (2004) 3215-‐3222. [112] R. Twombly, Avastin's uncertain future in breast cancer treatment, Journal of the National Cancer Institute, 103 (2011) 458-‐460. [113] E. Allen, I.B. Walters, D. Hanahan, Brivanib, a dual FGF/VEGF inhibitor, is active both first and second line against mouse pancreatic neuroendocrine tumors developing adaptive/evasive resistance to VEGF inhibition, Clinical cancer research : an official journal of the American Association for Cancer Research, 17 (2011) 5299-‐5310. [114] B.L. Falcon, K. Pietras, J. Chou, D. Chen, B. Sennino, D. Hanahan, D.M. McDonald, Increased vascular delivery and efficacy of chemotherapy after inhibition of platelet-‐derived growth factor-‐B, The American journal of pathology, 178 (2011) 2920-‐2930.
33
[115] J.T. Erler, K.L. Bennewith, T.R. Cox, G. Lang, D. Bird, A. Koong, Q.T. Le, A.J. Giaccia, Hypoxia-‐induced lysyl oxidase is a critical mediator of bone marrow cell recruitment to form the premetastatic niche, Cancer Cell, 15 (2009) 35-‐44. [116] C. Gialeli, A.D. Theocharis, N.K. Karamanos, Roles of matrix metalloproteinases in cancer progression and their pharmacological targeting, The FEBS journal, 278 (2011) 16-‐27. [117] E.S. Radisky, D.C. Radisky, Matrix metalloproteinase-‐induced epithelial-‐mesenchymal transition in breast cancer, Journal of mammary gland biology and neoplasia, 15 (2010) 201-‐212. [118] Y. Kojima, A. Acar, E.N. Eaton, K.T. Mellody, C. Scheel, I. Ben-‐Porath, T.T. Onder, Z.C. Wang, A.L. Richardson, R.A. Weinberg, A. Orimo, Autocrine TGF-‐beta and stromal cell-‐derived factor-‐1 (SDF-‐1) signaling drives the evolution of tumor-‐promoting mammary stromal myofibroblasts, Proceedings of the National Academy of Sciences of the United States of America, 107 (2010) 20009-‐20014. [119] U.M. Polanska, A. Acar, A. Orimo, Experimental generation of carcinoma-‐associated fibroblasts (CAFs) from human mammary fibroblasts, Journal of visualized experiments : JoVE, (2011) e3201.
Figure Legend
Figure 1. Histological images of sections from human tissue surgical specimens that
were immunohistochemically stained with some markers commonly used to identify
cancer-associated fibroblasts: alpha-smooth muscle actin (aSMA), fibroblast-specific
protein-1 (FSP-1), vimentin (VIM), platelet-derived growth factor receptor beta
(PDGFR-b); and pericytes: neuron glial antigen 2 (NG2). As discussed herein, no one
marker specifically labels all CAFs or clearly distinguishes CAFs from normal fibroblasts
or other closely related cell types. For example, aSMA not only denotes cancer-
associated fibroblasts (CAFs), but also myoepithelial cells (ME), pericytes (PC), and
smooth muscle cells (SMC) in both normal and cancer tissue. Shown is normal breast
tissue is from a female, age 23 (with the exception of the vimentin stain, which is from a
45 year old woman); normal prostate tissue is from a 51 year old man; cancerous breast
and prostate tumor tissues are from a 40 year old woman with ductal carcinoma and a 64
year old man with high grade adenocarcinoma, respectively. Positive staining for
indicated proteins appears brown. Tissue sections were counterstained with hematoxylin
34
(blue) to indicate cell nuclei and extracellular material. Scale bar = 100 µm. Images
were adapted and used with permission from The Human Protein Atlas
(www.proteinatlas.org) and only protein stainings with annotated expression ranked as
high reliability are displayed.